MITNET: a novel dataset and a two-stage deep learning approach for mitosis recognition in whole slide images of breast cancer tissue

Abstract

Mitosis assessment of breast cancer has a strong prognostic importance and is visually evaluated by pathologists. The inter, and intra-observer variability of this assessment is high. In this paper, a two-stage deep learning approach, named MITNET, has been applied to automatically detect nucleus and classify mitoses in whole slide images (WSI) of breast cancer. Moreover, this paper introduces two new datasets. The first dataset is used to detect the nucleus in the WSIs, which contains 139,124 annotated nuclei in 1749 patches extracted from 115 WSIs of breast cancer tissue, and the second dataset consists of 4908 mitotic cells and 4908 non-mitotic cells image samples extracted from 214 WSIs which is used for mitosis classification. The created datasets are used to train the MITNET network, which consists of two deep learning architectures, called MITNET-det and MITNET-rec, respectively, to isolate nuclei cells and identify the mitoses in WSIs. In MITNET-det architecture, to extract features from nucleus images and fuse them, CSPDarknet and Path Aggregation Network (PANet) are used, respectively, and then, a detection strategy using You Look Only Once (scaled-YOLOv4) is employed to detect nucleus at three different scales. In the classification part, the detected isolated nucleus images are passed through proposed MITNET-rec deep learning architecture, to identify the mitosis in the WSIs. Various deep learning classifiers and the proposed classifier are trained with a publicly available mitosis datasets (MIDOG and ATYPIA) and then, validated over our created dataset. The results verify that deep learning-based classifiers trained on MIDOG and ATYPIA have difficulties to recognize mitosis on our dataset which shows that the created mitosis dataset has unique features and characteristics. Besides this, the proposed classifier outperforms the state-of-the-art classifiers significantly and achieves a $$68.7\%$$ 68.7 % F1-score and $$49.0\%$$ 49.0 % F1-score on the MIDOG and the created mitosis datasets, respectively. Moreover, the experimental results reveal that the overall proposed MITNET framework detects the nucleus in WSIs with high detection rates and recognizes the mitotic cells in WSI with high F1-score which leads to the improvement of the accuracy of pathologists’ decision.

On the Analyses of Medical Images Using Traditional Machine Learning Techniques and Convolutional Neural Networks

“…Digital pathology has faced some significant obstacles, which has aided in the growth of computerized digital pathology methods. AMIDA 2013 Mitosis detection challenges sponsored by MICCAI Grand Challenge ICPR 2012 Contest on Mitosis Detection, EM segmentation challenge (ISBI - 2012) for the 2D segmentation of neuronal processes, GLAS for gland segmentation CAMELYON16 and TUPAC for processing breast cancer tissue samples are among the challenges that assessed both established and novel methods for analyzing digital pathology images [ 218 – 220 ].…”

Tubule-U-Net: a novel dataset and deep learning-based tubule segmentation framework in whole slide images of breast cancer

“…Breast cancer is the most frequent type of cancer in women, other than skin cancers, and is the leading cause of cancer death 1 , 2 . Early diagnosis of breast cancer is important and critical to raise the chance of successful treatment 2 .…”

“…Breast cancer is the most frequent type of cancer in women, other than skin cancers, and is the leading cause of cancer death 1 , 2 . Early diagnosis of breast cancer is important and critical to raise the chance of successful treatment 2 . For its assessment, Nottingham Histological Grading (NHG) system is generally used to measure the aggressiveness of breast cancer 3 , 4 .…”

“…Based on the NHG system 5 , there are three different significant factors that are used to obtain the overall scores of breast cancer. The NHG adds scores of tubule formation 6 , nuclear pleomorphism 7 , and mitotic count 2 , 8 , where each of which provides 1–3 points. Tubule formation is one of the most important assessment factors in the NHG grading for understanding the level of the cancer.…”

YOLOX-based framework for nuclei detection on whole-slide histopathological RGB and hyperspectral images

“…Automatic identification and classification of the different cells present in a digitalized sample would significantly reduce diagnosis time as well as pathologists' eyestrain. Previous works have studied the detection of mitosis cells on breast cancer histopathological RGB images 3,4 . These works were developed based on several competitions for mitosis breast cell recognition (ICPR12 5 , AMIDA13 6 , ICPR14 7 , and TUPAC16 8 ), offering high performance in this task.…”